JPH0736128B2 - Servo circuit speed detection gain automatic adjustment method - Google Patents

Description

DETAILED DESCRIPTION [Table of Contents] Outline Industrial field of application Conventional technology (FIGS. 7, 8, and 9) Problem to be solved by the invention Means for solving the problem (FIG. 1) ) Operation Example (a) Description of configuration of one example (FIGS. 2 and 3) (b) Description of operation of one example (FIGS. 4, 5, and 6) (c) Others Description of Embodiments of the Invention [Outline] A method for automatically adjusting a speed detection gain for speed control in a servo circuit that switches between speed control and position control to perform positioning control at a target position. A speed detection circuit that detects the actual speed from the position signal from the servo target and an error between the target speed and the actual speed are generated for the purpose of automatically adjusting the speed detection gain without any trouble. A speed error detection circuit and a position error signal is generated from the position signal. A position error detection circuit, a switching unit that switches and connects the servo target to the speed error detection circuit or the position error detection circuit, and a main control unit that generates a target speed and controls switching of the switching unit. Then, in the vicinity of the target position, in the servo circuit configured to switch from the speed control by the speed error detection circuit to the position control by the position error detection circuit, the differential gain of the speed detection circuit is changed to move the fixed distance. Repeatedly, a step of measuring the speed control continuation time in each differential gain with a counter, a step of obtaining an optimum differential gain of the speed control continuation time from the measured speed control continuation time, and a control current detection gain of the speed detection circuit It changes and repeats the movement of a certain distance,
The method has a step of measuring an integrated value of a position signal at least after position control in each control current detection gain, and a step of obtaining a control current detection gain of a minimum integrated value of the measured integrated values.

[Industrial application field]

The present invention relates to a method for automatically adjusting a speed detection gain for speed control in a servo circuit that switches between speed control and position control to control positioning at a target position.

The servo circuit is widely used for positioning control of a magnetic head of a magnetic disk device.

In such a servo circuit, it is necessary to adjust various gains, and in particular, there is a demand for a technique of appropriately adjusting the speed detection gain of the speed detection circuit in order to obtain an appropriate access time and an appropriate position control.

[Conventional technology]

 FIG. 7 and FIG. 8 are explanatory views of the prior art.

In FIG. 7, reference numeral 1 denotes a servo target, a voice coil motor 1a, a servo head 1b moved by the voice coil motor 1a, and a position signal generation circuit 1c for generating a position signal Ps from a read signal of the servo head 1b. have.

Reference numeral 2 denotes a speed detection circuit, which detects the actual speed Vr from the position signal Ps and a detection current ic described later. Reference numeral 3 denotes a speed error detection circuit, which is a speed error Δ between a target speed Vc and an actual speed Vr described later. V is generated and the speed is controlled.

A position error detection circuit 4 generates a position error signal ΔP from the position signal Ps and the detection current ic,
The position is controlled. A method of creating the position error signal ΔP is also disclosed in Japanese Patent Laid-Open No. 63-9084. Further, the detection current ic is added for the purpose of compensating for the phase delay of the position signal Ps.

Reference numeral 5 denotes a power amplifier and a changeover unit, which has a changeover switch and a power amplifier, and a speed error detection circuit 3 according to a coarse (speed control) / fine (position control) switching signal.
Alternatively, the position error detection circuit 4 is switched and connected to the servo target 1.

6 is a main control unit, which is composed of a microprocessor,
Generate a target speed curve Vc according to the amount of movement,
The position of the servo target 1 is monitored by a track crossing pulse described later, and a switching signal from coarse to fine is generated near the target position.

Reference numeral 7 is a control current detection circuit, which detects the control current Is of the power amplifier 5 and generates a detection current signal ic, and 8 is a track crossing pulse generation circuit, which is a position signal Ps.
Generates a track crossing pulse from the main control unit 6
Is output to.

When the number of moving tracks (moving amount) is given, the main control unit 6 generates a target speed curve Vc according to the number of moving tracks, and drives the voice coil motor 1a by speed control to reach the vicinity of the target position. , The switching unit 5 is switched to the position control side, the position of the voice coil motor 1a is controlled, and the voice coil motor 1a is positioned on a desired track.

As shown in FIG. 8, the speed detection circuit 2 detects the detection current.
An amplifier 20 for amplifying ic, a differentiating circuit 21 for differentiating the position signal Pc to generate a velocity component, and an offset adjusting circuit 22.
Further, it has an amplifier 23 for adding and amplifying these outputs, and it is possible to perform an initial adjustment of the control current detection gain and the differential gain by operating the variable resistors r ₁ and r ₂ .

That is, the speed detection circuit 2 differentiates a known position signal,
This is a circuit that obtains the actual velocity signal, and a detection current ic is added to compensate for the phase delay of the position signal Ps.

The adjustment of the control current detection gain and the differential gain, which are the related speed detection gains, is performed by moving (seeking) while observing the position signal Ps with an oscilloscope to adjust the variable resistance r ₁
Was adjusted to adjust the control current gain so that the overshoot and undershoot of the position signal Ps shown in FIG.

Also, for the adjustment of the differential gain, perform seek operation while observing the coarse / fine switching signal with an oscilloscope,
The variable resistance r ₂ was adjusted so that the coarse (speed control) time tc in FIG. 9 (A) was within a predetermined range.

[Problems to be Solved by the Invention]

However, in the conventional technique, since a human uses an oscilloscope to adjust the variable resistance, there is a problem that an adjustment error is likely to occur due to an individual difference or an error in a measuring instrument. Therefore, there is also a problem that the adjustment cost becomes large.

Therefore, the present invention, without messing with human hands,
An object of the present invention is to provide a method for automatically adjusting a speed detection gain of a servo circuit, which is capable of automatically adjusting the speed detection gain.

[Means for Solving the Problems]

 FIG. 1 is a principle diagram of the present invention.

The present invention, as shown in FIG. 1, includes a speed detection circuit 2 for detecting an actual speed from a position signal from a servo target 1, and a speed error detection circuit 3 for generating an error between a target speed and the actual speed.
A position error detection circuit 4 for generating a position error signal from the position signal, and a switching unit 5 for switching and connecting the servo object 1 to the speed error detection circuit 3 or the position error detection circuit 4.
And a main control unit 6 that generates a target speed and controls the switching of the switching unit 5. In the vicinity of the target position, from the speed control by the speed error detection circuit 3 to the position control by the position error detection circuit 4. In the servo circuit configured to switch to, the step of measuring the speed control continuation time at each differential gain by changing the differential gain of the speed detection circuit 2 and repeating the movement of a constant distance with a counter, and the measured speed control. A step of obtaining an optimum differential gain of the speed control continuation time from the continuation time, a control current detection gain of the speed detection circuit 2 is changed, and a movement of a constant distance is repeated, and at least position control after each control current detection gain is performed. Measuring the integral value of the position signal of, and obtaining the control current detection gain of the minimum integral value of the measured integral values. Those having.

[Action]

In the present invention, since the access time (speed control continuation time) changes depending on the differential gain, the differential gain is changed, the speed control continuation time at each differential gain is measured by the counter, and the optimum differential gain of the speed control continuation time is obtained. Ask for.

Next, the control current detection gain changes the waveform of the position signal Ps before the coarse / fine switching, and affects the position error signal ΔP after the fine control.

That is, as for the actual speed Vr, the lower the seek speed becomes, the greater the ratio of correction of the detected current becomes. Therefore, at the end of the seek, the waveform of the position signal at the end of the seek can be changed by changing the control current detection gain that determines the correction amount of the detected current.

Since it is desirable that the position signal Ps converge to 0 immediately after the start of the fine control, the position signal Ps is integrated to obtain the control error, and the control current detection gain that minimizes the integrated value is obtained. The optimization of the waveform of the position signal Ps that minimizes is realized.

〔Example〕

(A) Description of Configuration of One Embodiment FIG. 2 is a configuration diagram of an embodiment of the present invention, and FIG. 3 is a configuration diagram of a speed detection circuit in the configuration of FIG.

In the figure, the same components as those shown in FIGS. 1, 7, and 8 are designated by the same symbols.

Reference numeral 9a is a counter, which is started / stopped by the main control unit 6 to measure the speed control duration tc, and 9b is an integration unit, which is a main control unit (hereinafter referred to as MPU) 6
Switch that is turned on by the position signal Ps to pass
90, an absolute value circuit 91 for converting the position signal Ps from the switch 90 into an absolute value, and an integrating circuit 92 for integrating the output of the absolute value circuit 91.
And an analog / digital converter (ADC) 93 for converting the analog output of the integrating circuit 92 into a digital value.

Reference numerals 60 and 61 are gain registers, the gain register 60 is for storing the control current detection gain M, the gain register 61 is for storing the differential gain N, and 62 is a flag register for controlling the adjustment processing. What is used, I is an integration counter, which stores the number of integrations, 64 is a work register, and various measured values Mf, Mr,
It stores T ₁ , T ₂ , T ₃ , and A.

In FIG. 3, 24 and 25 are multiplication type digital / analog converters (referred to as D / A converters), each of which has an amplifier 2
The detection current ic from 0 is multiplied by the control current detection gain M of the main control unit 6 and the speed signal of the differentiating circuit 21 is multiplied by the differential gain N of the main control unit 6 to output the result.

(B) Description of operation of one embodiment FIG. 4 is a flow chart of adjustment processing of one embodiment of the present invention, FIG. 5 is a flow chart of overshoot / undershoot adjustment processing in FIG. 4, and FIG. 6 is FIG. 6 is a flow chart of integral sampling processing in FIG.

FIG. 6 is the subroutine of FIG. 5, and FIG. 5 is the subroutine of FIG.

First, the flow of FIG. 4 will be described.

At the start of adjustment, the MPU 6 sets the adjusted flag F to "0" and initializes various registers.

At this time, initial values "M" and "N" are set in the gain registers 60 and 61 respectively, and the gain related to the speed detection circuit 2 is given.

The MPU 6 seeks (moves) the voice coil motor 1a to the scheduled start point.

When the movement to the scheduled start point is completed, the MPU 6 resets the counter 9a to measure the access time, and starts after resetting.

Then, the MPU 6 starts a seek for a predetermined difference amount d from this start point.

Therefore, the speed of the voice coil motor 1a is controlled by the speed error detection circuit 3.

The MPU 6 counts the track crossing pulses of the track crossing pulse generation circuit 8, and when it detects that it has reached the vicinity of the target position, ends the speed control and switches to the position control.

Along with this, the counter 9a is stopped.

As a result, the counter 9a measures the access time (speed control continuation time) tc.

Then, the MPU 6 converges to the target position by the position control when the on-track signal of the position error detection circuit 4 (the signal output when the position error signal ΔP is within a certain range) continues for a certain time (800 μs). And it is judged that the seek has ended.

Next, the MPU 6 reads the measured value of the counter 9a and checks whether the measured value of the counter 9a is within the expected range.

If it is within the predetermined range, the control current detection gain of the step is adjusted. If it is not within the predetermined target range, the differential gain is adjusted, and the step is advanced.

If it is not within the planned target range, the adjusted flag F is reset to "0" for re-adjustment.

If the measured value of counter 9a is earlier than the target, register 61
The differential gain N is increased to (N + 1), and if it is not faster than the target, the differential gain N is reduced to (N-1), and the differential gain N is output to the DAC 25 of the speed detection circuit 2, and the process returns to step.

That is, if it is faster than the target, the differential gain is set to a large value and the actual speed Vr
To make the access time slower, and if it is not faster than the target, the differential gain is made small and the actual speed Vr is made small to make the access time fast.

On the other hand, if the measured value of the counter 9b is within the expected range, the MPU 6 checks the adjusted flag F, and if it indicates that overshoot / undershoot of F = “1” has been adjusted, the process ends.

On the contrary, if F = “1”, that is, if F = “0”, the overshoot / undershoot adjustment has not been completed.
In the undershoot adjustment subroutine, the adjustment gain Mf in the forward seek direction is obtained and stored in the register 64.

Next, the MPU 6 obtains the adjustment gain Mr in the reverse seek direction by the overshoot / undershoot adjustment subroutine described later in FIG. 5, and stores it in the register 64.

Further, the average of the forward seek adjustment gain Mf and the reverse seek adjustment gain Mr is calculated, stored in the register 60 as the control current detection gain M, and the process returns to the step.

Next, the overshoot / undershoot adjustment processing will be described with reference to FIG.

First, the MUP 6 sets “3” in the integrating circuit I of the register 63. That is, the integration is performed 3 times.

The MUP6 outputs the control current detection gain M of the register 60 to the DAC 24 of the speed detection circuit 2.

Then, the MUP 6 executes the integral sampling subroutine described later with reference to FIG. 6, obtains the integral value of the position signal Ps in the register A, and stores it in the register Ti (i = 4-I).

At this time, this routine is repeated a plurality of times to average the integrated values.

Next, the MUP 6 updates the gain M of the register 60 to (M + X) and the number of integrations I of the register 63 to (I-1). However, X is a predetermined displacement amount. This value is not a value directly related to the overshoot / undershoot adjustment processing, and is only necessary for changing the control current detection gain at equal intervals.

The MUP 6 checks whether the integration count I of the register 63 is "0", and if it is not "0", returns to the step.

On the other hand, if I = 0, the three integration operations have ended, and the integrated values T ₁ , T ₂ , and T ₃ have been obtained, and the current gain is (M
+ 3X).

First, the MUP 6 compares the first-time integrated value T ₁ with the second-time integrated value T ₂ .

If T ₁ ≧ T ₂ is not satisfied, that is, T ₁ <T _2, then a minimal value cannot be obtained because of a monotonic increase with respect to an increase change in the gain M. Therefore, the gain M is (M−4X), that is, M = M + 3X. Therefore, reduce to (MX) and return to the step.

On the other hand, if T ₁ ≧ T ₂ , the second integrated value T ₂ and the third integrated value T ₃ are compared.

If T ₃ ≧ T ₂ is not satisfied, that is, T ₃ <T ₂ , the gain M is (M−2X), that is, (M−X), because the minimum value cannot be obtained because the gain M monotonically decreases as the gain M changes. And increase to step.

On the contrary, if T ₃ ≧ T ₂ , the relationship of T ₁ ≧ T ₂ ≦ T ₃ is established, and T ₂ ≧ T ₂
Is a minimum value, the gain of T ₂ is (M−2X) = (M +
X) and store it in the register 64 as the forward control current detection gain Mf, and set the adjusted flag F to "1".
Set to and return.

Incidentally, the control current detection gain Mr in the reverse direction is similarly obtained by performing integral sampling in the reverse direction in steps.

Next, the integral sampling process will be described with reference to FIG.

(I) The MPU 6 starts forward seek for the planned amount of difference.

(Ii) The MPU 6 judges whether it is half track before the target position, and when it is half track before, it issues an integration start, turns on the switch 90, and operates the integration circuit 92.

Therefore, the integrator circuit 92 starts integrating the position signal Ps from a half track before, as shown in FIG.

(Iii) After that, the speed control is switched to the position control, and the seek ends when the on-track signal continues for a certain time as in the step of FIG.

After waiting a predetermined time, the integration start signal is turned off, the switch 90 is turned off, the integration circuit 92 is made inoperative, and the integration is completed.

Therefore, the integration period is as shown in FIG.

(Iv) After the integration period ends, the MPU 6 samples the integrated value from the ADC 93 and stores it in the register 64 as “A”.

Then, a reverse seek is performed by the predetermined amount and the process returns.

The above flow is the integral sampling process in the forward direction, but in the case of the reverse direction, the forward seek is set to the reverse seek in step (i), and the step (iv) is performed.
Just change reverse seek to forward seek with
The rest is the same.

In this way, in FIG. 4, the differential gain N for the proper access time is obtained, and then the control current detection gain M for the proper positioning waveform is obtained.

Since the access time occupies most of the positioning time, first, the differential gain for the optimization of the access time is adjusted, and the control current detection gain N that makes the minimum overshoot / undershoot in the differential gain is obtained.

Furthermore, by changing the control current detection gain N, the differential gain is adjusted again in order to confirm whether the access time is within the predetermined range.

Then, if the access time is out of the predetermined range, the differential gain is adjusted again.

Also, the reason why the integration for adjusting the control current detection gain is performed from before the half track is that the control current detection gain affects the coarse control (speed control), and the position signal Ps at the coarse / fine switching is zero. This is because the plunge angle into the bolt affects the undershoot and overshoot of the subsequent position control.

Therefore, the integration is performed from before the half track, that is, including the coarse period immediately before the coarse / fine switching.

(C) Description of Other Embodiments In the above-mentioned embodiment, the integration is performed from before the half track. However, if the position signal in the position control is integrated, the effect of the gain on the plunge angle can be understood to some extent. The position signal may be integrated from the start of position control.

Further, although the magnetic disk device has been described as an example, another device may be used, and the operation of the counter 9a and / or the integrating unit 9b is controlled by MP.
U6 may execute.

Although the present invention has been described with reference to the embodiments, the present invention can be variously modified according to the gist of the present invention, and these modifications are not excluded from the present invention.

〔The invention's effect〕

As described above, the present invention has the following effects.

Since the differential gain and the control current detection gain are changed as the speed detection gain and automatically adjusted, the seek speed can be automatically adjusted while maintaining high stability at the end of the seek.

Seek speed is measured by repeatedly seeking a certain amount of movement and counting the speed control time.
The degree increases.

By changing the control current detection gain and measuring the settling characteristic at the end of seek (the amount of overshoot / undershoot at the end of seek) from the integrated value of the position signal, the stability at the end of seek can be adjusted. The stability of is higher.

[Brief description of drawings]

FIG. 1 is a principle diagram of the present invention, FIG. 2 is a configuration diagram of an embodiment of the present invention, FIG. 3 is a configuration diagram of a speed detection circuit in the configuration of FIG. 2, and FIG. 4 is an adjustment of an embodiment of the present invention. Processing flow chart, FIG. 5 is a flow chart of overshoot / undershoot adjustment processing in FIG. 4, FIG. 6 is a flow chart of integral sampling processing in FIG. 5, and FIGS. 7, 8 and 9 are prior arts. FIG. In the figure, 1 ... Servo object, 2 ... Speed detection circuit, 3 ... Speed error detection circuit, 4 ... Position error detection circuit, 5 ... Switching unit, 6 ... Main control unit.

Continuation of the front page (51) Int.Cl. ⁶ Identification number Office reference number FI technical display location H02P 5/00 101 E 8625-5H

Claims (1)

[Claims] 1. A speed detection circuit (2) for detecting an actual speed from a position signal from a servo target (1), a speed error detection circuit (3) for generating an error between a target speed and the actual speed, A position error detection circuit (4) for generating a position error signal from a position signal, and a switching unit (5) for switching and connecting the servo target (1) to the speed error detection circuit (3) or the position error detection circuit (4). ) And a main control unit (6) that generates a target speed and controls the switching of the switching unit (5). In the vicinity of the target position, the speed error detection circuit (3) controls the position In the servo circuit adapted to switch to the position control by the error detection circuit (4), the differential gain of the speed detection circuit (2) is changed and the movement of a fixed distance is repeated, and the speed control continuation time at each differential gain is counted. so The step of measuring, the step of obtaining an optimum differential gain of the speed control duration from the measured speed control duration, the control current detection gain of the speed detection circuit (2) is changed, and the movement of a constant distance is repeated. The step of measuring the integrated value of at least the position signal after the position control in each control current detection gain, and the step of obtaining the control current detection gain of the minimum integrated value of the measured integrated values. Servo circuit speed detection gain automatic adjustment method.